r8 is a simple jeopardy-style CTF system. What sets it apart from other platforms?
- r8 is intentionally simple. It won't send emails, support multi-server deployments, or provide an LDAP integration.
- r8 is designed to support university courses. It can be deployed for an entire semester and includes challenge scheduling functionality and logging capabilities to detect cheating.
- r8 is written in modern Python 3. This generally makes it easy to spawn additional network services or interface with other tools and languages.
r8 is successfully being used for teaching at the University of California, Berkeley and the University of Innsbruck, Austria.
In short, install r8 as a Python package using your preferred way. We recommend the following:
Make sure you have Python 3.9 or above. Clone the repository, create a Python virtual environment into which we install all dependencies, and finally install r8:
python3 -m venv venv git clone https://github.com/mhils/r8.git venv/bin/pip install -e ./r8
Activate the virtual environment. This always needs to be done to make the
r8 command available:
Create r8's SQLite database in the current directory. We also need to let r8 know under which origin it will be hosted (on which hostname/URL the web interface will be served) at this step:
r8 sql init --origin http://localhost:8000
r8 is typically configured with a plain SQL file. Let's add some demo challenges and users:
r8 sql file r8/config.sql
We can now start r8:
You can now browse to http://localhost:8000/ and log in as
user1 with password
r8has a comprehensive command line interface. Check out
r8 users --help, etc.
- Take a look at
config.sqlto learn how r8 can be configured.
- Install additional challenges and create new ones (see next section).
Installing additional challenges
We likely want to install additional challenges, for example from the r8-example repository. To make challenges available to r8, we need to install the corresponding Python package into our Python environment. Let's get the example repository and add it:
git clone https://github.com/mhils/r8-example.git venv/bin/pip install -e r8-example
We can now verify that r8 has picked up the new challenges:
r8 challenges list-available # Output: # r8.builtin_challenges: # [...] # r8_example: # - HelloWorld
To make the challenge available to users, we also need to instantiate it by adding it to the database.
Go back to the SQL configuration file (
config.sql in the example above) and add the following:
INSERT INTO challenges (cid, team, t_start, t_stop) VALUES ('HelloWorld', 0, datetime('now'), datetime('now','+1 month'));
Finally, we can apply our configuration changes and run r8 again:
r8 sql file config.sql r8 run
The Hello World challenge is now visible to users!
Creating new challenges
The API Documentation for challenge development can be found at https://mhils.github.io/r8/.
It is recommended to use r8-example as a template
and place challenges in a new separate repository. See
for challenge examples.
Customizing the HTML template
r8 provides some simple means to modify the default HTML template, for example to add a custom logo.
An example can be found in the
r8 consists of the following parts:
- The core
r8application written in Python, which manages the currently active challenges. It provides a command-line API for administration (
r8.cli), a REST API for users (
r8.server), and a Python API for challenges (
- CTF challenges implemented in Python 3.8+. All challenges need to inherit from
r8.Challengeand must be registered using entrypoints so that they are imported on start. See
r8_examplefor challenge examples and each repo's
setup.pyfor entrypoint declaration.
- An SQLite database that contains information on users, groups, challenge scheduling, and flags. There also is an event log that can be used to e.g. detect indicators of plagiarism.
- A web interface that allows users to view challenges and enter flags, implemented using React and Bootstrap. To simplify development, there is no compilation step.
To speed up development, the server can be automatically reloaded on changes using modd.
For production use, it is recommended to run r8 on a throwaway VM behind a TLS-terminating reverse proxy such as nginx. A couple of auxiliary configuration examples are provided in the ./misc folder:
crontab: cronjob to make daily backups.
nginx.conf: nginx configuration example for an HTTPS-only deployment.
r8.service: systemd service file example.